October 15, 2018
Introduction
Team
Samantha Sprague
- Principal Flavor Scientist
- Creative Genius
David Josephs
- Senior Hopps Engineer
- Data Janitor
Carson Drake
- Lead Taste Data Analyst
- Code Coordinator
QOTD: Given the datasets Beer and Brews, we want to know:
-Where our headquarters should be?
-What style of beer would appeal to the widest audience?
-If we wanted to have a high ABV, what is the ideal IBU?
The Data
| BeerName | Beer_ID | ABV | IBU | Style | Ounces | BreweryName | Brewery_ID | City | State |
|---|---|---|---|---|---|---|---|---|---|
| Get Together | 2692 | 0.045 | 50 | American IPA | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
| Maggie's Leap | 2691 | 0.049 | 26 | Milk / Sweet Stout | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
| Wall's End | 2690 | 0.048 | 19 | English Brown Ale | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
| Pumpion | 2689 | 0.060 | 38 | Pumpkin Ale | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
| Stronghold | 2688 | 0.060 | 25 | American Porter | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
| Parapet ESB | 2687 | 0.056 | 47 | Extra Special / Strong Bitter (ESB) | 16 | NorthGate Brewing | 1 | Minneapolis | MN |
Caption: Here is a quick glance at the head of our merged dataset. You can see we created easy to read captions and merged the two raw datasets with our renames field "Brewery_ID".
Here are the first 6 rows of our merged data set.
The Data (continued)
| BeerName | Beer_ID | ABV | IBU | Style | Ounces | BreweryName | Brewery_ID | City | State | |
|---|---|---|---|---|---|---|---|---|---|---|
| 2405 | Pilsner Ukiah | 98 | 0.055 | NA | German Pilsener | 12 | Ukiah Brewing Company | 556 | Ukiah | CA |
| 2406 | Heinnieweisse Weissebier | 52 | 0.049 | NA | Hefeweizen | 12 | Butternuts Beer and Ale | 557 | Garrattsville | NY |
| 2407 | Snapperhead IPA | 51 | 0.068 | NA | American IPA | 12 | Butternuts Beer and Ale | 557 | Garrattsville | NY |
| 2408 | Moo Thunder Stout | 50 | 0.049 | NA | Milk / Sweet Stout | 12 | Butternuts Beer and Ale | 557 | Garrattsville | NY |
| 2409 | Porkslap Pale Ale | 49 | 0.043 | NA | American Pale Ale (APA) | 12 | Butternuts Beer and Ale | 557 | Garrattsville | NY |
| 2410 | Urban Wilderness Pale Ale | 30 | 0.049 | NA | English Pale Ale | 12 | Sleeping Lady Brewing Company | 558 | Anchorage | AK |
Caption: Here is a quick glance at the tail of our merged dataset. You can see we created easy to read captions and merged the two raw datasets with our renames field "Brewery_ID".
Here are the last 6 rows of our merged data set.
The Data: Median Values and Breweries by State
| State | medianABV | medianIBU | NumBreweries |
|---|---|---|---|
| AK | 0.0560 | 46.0 | 7 |
| AL | 0.0600 | 43.0 | 3 |
| AR | 0.0520 | 39.0 | 2 |
| AZ | 0.0550 | 20.5 | 11 |
| CA | 0.0580 | 42.0 | 39 |
| CO | 0.0605 | 40.0 | 47 |
In order to make the dataset more friendly for analysis, we created a new dataset from our merged dataset that would look at the median ABVs and IBUs, along with the number of breweries for each state.
Missing Values
The next logical step in analyzing the dataset was to find missing values.
na_count BeerName 0 Beer_ID 0 ABV 62 IBU 1005 Style 0 Ounces 0 BreweryName 0 Brewery_ID 0 City 0 State 0
There are 62 missing data points for ABV, and 1005 missing data points for IBU.
Graphics: Number of Breweries
We can see that the Colorado has the darkest coloring on the graph which alludes to the highest number of breweries! Colorado contains 47 breweries. California, Oregon, Texas and Washington fall behind with 39, 29, 28, and 23 breweries.
Graphics: Analysis on ABV
Median ABV
ABV Distribution
Maximum ABV
| State | BeerName | City | ABV |
|---|---|---|---|
| CO | Lee Hill Series Vol. 5 - Belgian Style Quadrupel Ale | Boulder | 0.128 |
Colorado has the most alcoholic beer of any state! Lee Hill Series Vol. 5, a Belgian Style Quadrupel Ale produced in Boulder,CO has an ABV of 12.8%. We also see that Kentucky and Washington DC prefer the most alcoholic beers overall, while in general, the states prefer beer with an average alcoholic content, giving the distribution a tight, somewhat normal shape.
Graphics: Analysis on IBU
Median IBU
IBU Distribution
Maximum IBU
| State | BeerName | City | IBU |
|---|---|---|---|
| OR | Bitter Bitch Imperial IPA | Astoria | 138 |
Oregon has the bitterest beer of any state! Bitter Bitch Imperial IPA, produced in Astoria, has an IBU of 138! We also see that overall, Maine has the bitterest beers. It seems that most states prefer beer which is less bitter than average, giving the distribution of bitterness a right skew.
Graphics: Analysis on ABV vs IBU
ABV v IBU
Another view: ABV vs IBU
Another view: Style Analysis
Numbers: Summary Statistics
Summary Statistics for ABV
Min. 1st Qu. Median Mean 3rd Qu. Max. NA's 0.00100 0.05000 0.05600 0.05977 0.06700 0.12800 62
You can see that the median ABV for all beers is roughly 0.056 and the mean ABV is 0.5977. It is also worth noting that there at 62 beers who do not have a recorded ABV. Those beers were not considered in the analysis for ABV.
Summary Statistics for IBU
Min. 1st Qu. Median Mean 3rd Qu. Max. NA's 4.00 21.00 35.00 42.71 64.00 138.00 1005
You can see that the median IBU for all beers is roughly 35 and the mean IBU is 42.71. It is also worth noting that there at 1005 beers who do not have a recorded IBU. Those beers were not considered in the analysis for IBU.
Conclusions
Headquarter and Style of Beer Recommendation for Machine Lite
We recommend:
Target Market: Arkansas, Mississippi, and Alabama
State Headquarter: Mississippi
According to our linear model, for an above average ABV beer at 0.06 (mean= 0.05977), the ideal IBU is roughly 35-40.
According to our density plots and style treemap, the style of beer which would appeal the wides audiance is the American IPA.
Thank you
Questions?
